Problems in automated recognition of valley features from digital elevation models and a new method toward their resolution

Publisher Website
Google Scholar
Abstract
Valley recognition methods given in the literature fail to produce valley networks suitable for locating valley heads for one of two main reasons: (a) they are based on the concept of [higher than] or (b) they are based on contributing area thresholds. In this paper a new method for recognizing valley bottoms is presented. This is based on two improvements of Carroll's (1983) method, and produces a network that reflects the topography well. The network is used to locate valley heads. Once located, the valley heads are delineated using criteria suggested by geomorphologists and hydrologists. The resultant valley heads are generally well recognized although two problems are evident: (a) there are a number of commission errors; and (b) the valley heads recognized by the method are too small.

This publication has 12 references indexed in Scilit: